Project Summary Dexterous forelimb tasks vary in their degree of difficulty, but whether someone is performing a life-saving surgery or reaching for a cup of coffee, the neural mechanisms that underlie these goal-directed movements are poorly understood. Previous studies have shown that as people perform dexterous forelimb tasks, their limb movement is constantly being updated as it traverses through space toward a target. This consistent and rapid updating of forelimb movement suggests that feedback is critical for the success of goal-directed forelimb movements. However, rapid corrections are performed too quickly to be explained purely by sensory feedback, because of delays in the transmission of sensory signals from the body to the nervous system. Neuroscientists have turned to engineering principles of control theory to generate hypotheses regarding the biological elements of motor control. One theory with considerable empirical evidence is that during forelimb movement, internally directed efference copies are sent to the cerebellum, so that the cerebellum can make predictions about the subsequent state of the limb. Generally, evidence for efference copies has been relegated to electrophysiology and/or behavior studies that demonstrate motor output modulation arriving faster than sensory feedback, but the characterization of putative neural structures is lacking. The dearth of evidence characterizing putative neural structures as either sending, processing, or receiving efference copies is due to the difficulties in accessing these structures. However, the use of molecular-genetic tools in the mouse, to selectively access specific cell types within a neural structure, has made potential efference copy circuits more accessible. In order to understand how the cerebellum is using efference copies to generate predictions about subsequent forelimb positions, it is imperative that the pre-cerebellar processing and organization of efference copies are elucidated. The lateral reticular nucleus (LRN) is a pre-cerebellar structure entirely composed of neurons that send mossy fiber projections to the cerebellum. A subset of the input to the LRN comes from the ascending branch of cervical propriospinal interneurons (PNs). PNs are characterized as spinal interneurons receiving descending motor commands and sending bifurcating axonal projections; one branch descending to forelimb motor neurons, and the other branch ascending to the LRN, carrying efference copy information. Therefore, the LRN is an optimal target for evaluating the role of efference copies in skilled forelimb movement. This proposal attempts to investigate the functional role of the LRN in skilled forelimb movement by 1) perturbing LRN function during skilled forelimb movement, and 2) using multi-channel in vivo extracellular recordings to characterize efference copy organization and processing in the LRN. Beyond dissecting motor control circuits, these findings may help improve diagnostics and therapeutics for individuals suffering from devastating motor and neurologic deficits. These findings may also provide strategies for designing and improving neural prostheses.